The French playwright Moliere loved poking fun at pomposity especially when it involved doctors. In Le Bourgeois Gentilhomme, he pokes fun at Monsieur Jourdain, a pompous upstart, who has come into money and is taking on airs and graces.
Translating gentilhomme into gentleman misses what the word meant around 1660, which was more like civilized, the manners which aristocrats, our betters, had in contrast to which the rest of us were savages, wild, uncivilized or uncouth – something more like grossier in French, gross in English slang.
Savage in English is sauvage in French, which translates back to English as wild.
Words count in a black comedy or perhaps horror story. Doctors have done a reverse Jourdain – gone from being civilized to médecins sauvage or grossier – take your pick.
Could such a tale be possible? It was written in 1886 – The Strange Case of Dr. Jekyll and Mr. Hyde.
The Degenerate Doctor
Jekyll and Hyde is one of the most famous books in English and the phrase Jeykll and Hyde is widely used, even by the French. Psychiatrists claim the story illustrates multiple personality disorder or substance abuse, or under the influence of Freud that it speaks to our unconscious desires. But it was written before Freud or the split personality epidemic.
The dominant idea in 1886 was Cesare Lombroso’s L’Uomo Deliquente – The Degenerate Man. Lombroso’s vision was that some people were savages without a conscience, beyond the reach of religion or education, and through misbreeding we all might regress to savagery.
Henry Jekyll was the epitome of a civilised man – a doctor and scientist researching drugs. In a marvellous 1941 movie, Spencer Tracy played Jekyll – exactly the kind of family doctor you’d like to have, the kind some of us can remember from childhood.
A drug produces his transformation – into a restless, disinhibited, homicidal creature. An SSRI? It is possible serotonin shaped the story – Stevenson did take cocaine.
Poisons for Peace
The long decade from 1945 to 1962 was a miraculous one for drug discovery. We got steroids, contraceptives, antibiotics, antihypertensives, hypogylcemics, antipsychotics, antidepressants, tranquilizers, cancer chemotherapies and more. Some of these remain the best drugs in their class we have ever had.
This decade cemented a belief in science in most of us – it was becoming possible to eliminate scourges that had plagued us for millennia, robbing parents of more than half their children in childhood, killing great artists, scientists and others prematurely.
Doctors were civilised in the 1950s. They didn’t just discover new drugs, they were alert to their hazards. In 1952, Albe Watkins established that chloramphenicol causes aplastic anemia. Frank Ayd reported tardive dyskinesia on antipsychotics and he and others reported on the sexual, suicidal and dependence producing effects of antidepressants. Leo Meyler wrote the first book on The Side Effects of Drugs. And in 1960, most remarkably given that the side effect could not be seen happening in front of a doctor, Widukind Lenz proved thalidomide caused birth defects.
Louis Lasagna was among the most decent and civilised. A rising star, his name already on one of the most cited medical papers of all time – about placebos. At great cost to himself, he stood by the view that early oral contraceptives could cause blood clots.
He was among the first clinical pharmacologists, working on the science of therapeutics, one part of which is now called pharmacovigilance – the science of establishing whether Drug X can cause Problem Y. Pharmacovigilance was created and driven by doctors, reporting their cause and effect judgements in journals like the New England Journal of Medicine and the British Medical Journal.
Lasagna was also the first proponent of Randomized Controlled Trials (RCTs) in America. Tony Hill had run the first RCT in Britain in 1947. It was quicker than the usual evaluations of drugs doctors did but could miss important details. In 1956, Lasagna suggested drug companies should be required not just to show their drugs were safe but also that they worked – given RCTs now offered a simple and quick way to do this.
No-one paid any heed. Up till then the regulation of drugs, under the 1938 Food and Drugs Act, like all regulation, whether food or automobiles or stock markets, had been about keeping us safe. Besides none of the obviously effective wonder drugs down flooding into doctor’s offices had been through a controlled trial of any sort before coming on the market.
A Tragic Over-Reach
The unimaginably horrific birth defects thalidomide brought in its wake only reached America in 1962. Arguing a sleeping pill should not cause a peripheral neuropathy, using the 1938 Act, Frances Kelsey kept thalidomide off the US market.
Nonetheless, something had to be seen to be done. In 1962, a new FDA Act forced companies to show their drugs were both Effective and Safe. RCTs were brought in as the quick and simple way to do this. This extra constraint on companies would keep us all safe. It occurred to very few – but did occur to some – that handing the responsibility to prove a drug worked over to companies might not be a good idea. It occurred to no-one that the same company RCTs might also shape how we viewed safety.
There was a prior indication things might not turn out as hoped. In 1960, a Lasagna placebo controlled RCT showed thalidomide was an effective sleeping pill – but missed the agitation, suicidality, sexual dysfunction, peripheral neuropathy and birth defects it causes. When Marilyn Monroe died of a sleeping pill overdose, Lasagna remarked she would be alive today if she had been taking thalidomide. Two months later, the new FDA Act would have made it impossible to stop it being licensed.
What does an Effective drug look like? Most of us would say it should save lives and reduce disability. An RCT demonstrating that would take an exceptional national effort among doctors or anything between 5 to 10 years if run by a company. License applications for most drugs may now be technically illegal in that we have defaulted from Effective to Effect – companies just need to show an Effect on lipid levels or rating scale scores. Placebos let them get drugs with very weak effects on the market.
Many drugs now marketed as working (having an Effect) kill and disable more than they save – very few drugs in recent decades show any evidence of lives saved.
Lecturing doctors about RCTs in 1965, Tony Hill remarked that the people you hear encouraging doctors to prescribe according to the latest clinical trial mostly work for pharmaceutical companies. RCTs only tell us about the average effects of drugs not how to treat the person in front of us. They miss side-effects. Your judgement about what a drug does, good or bad, he told them is more important than randomization, blinding or placebos. Forget this and we open the door to medical lunacy.
In 1983, Lasagna was faced with experts saying that RCTs were not just the best way to establish if a drug worked but also what adverse effects it might have.
Reporting [by doctors] is “the least sophisticated and scientifically rigorous . . . method of detecting new adverse drug reactions”
This claim is completely wrong. Drugs do hundreds of things simultaneously but RCTs are designed to look at one thing only. Lasagna responded to the claim:
This may be true in the dictionary sense of sophisticated meaning ‘adulterated’ . . . but I submit reporting [by doctors] is more worldly-wise, knowing, subtle and intellectually appealing than… RCTs.
He is playing on words. Sophisticating wine means adulterating it.
Some years later, Lasagna, looking at effectiveness, echoed Hill saying:
Evidence Based Medicine… RCTs… invariably fail to tell the physician what he or she wants to know which is which drug is best for Mr Jones or Ms Smith – not what happens to a non-existent average person
The Dram of Eale
“The dram of eale doth all the noble substance often dout to his own scandal”
The trace of a fault (evil) in a man often casts doubt on what is good in him leading to his downfall.
This line from Hamlet is the Shakespearean definition of tragedy. What was the element of eale that transformed doctors into savages with all the problems that has brought us?
It wasn’t a drug. It wasn’t Lasagna. It was another man with good intentions.
Seven years after Lasagna said that good clinical interviews were the way to recognize a drug was causing a problem, in a 1990 article 3 Boston clinicians claimed fluoxetine caused 6 people to become suicidal. Analyzing the cases closely using good clinical science to link cause and effect, this article established that fluoxetine could make some people suicidal.
Other groups reported similar findings. Some patients were challenged, dechallenged and rechallenged with an SSRI. Ditto for children. There was no way to explain what happened except that fluoxetine had caused it. This was Evident Based Medicine of the type doctors like Lasagna had been practising since the 1950s – it was and is supremely solid science.
In response, BMJ published an article, Beasley et al, in which Eli Lilly, the makers of Prozac, fluoxetine, claimed their analysis of their RCTs showed no evidence Prozac made anyone suicidal. The case reports, they said, were sad but anecdotal – and the plural of anecdote is not data. Depression was the problem not fluoxetine. Clinical trials are the science of cause and effect. Doctors, the media, politicians and all of us were challenged to choose between believing the science or the anecdotes.
FDA and other SSRI companies had input into this article that created Evidence Based Medicine (EBM) and just as with RCTs 30 years earlier, its pharma companies who most commonly exhort doctors to practice EBM today.
But the original phrase was the plural of anecdotes is data – otherwise Google wouldn’t work.
It’s difficult to see how depression might have caused suicide attempts and suicides in company healthy volunteer trials but companies wheel out experts to say just that.
The key point is, the case reports, Evident Based Medicine, is the science. In so far as it contains company trials, EBM is an artefact. Which are you going to believe the Science or the Artefact? The Science of Medicine lies in making hard judgement calls. The made by algorithm approach, combined with inappropriate statistics, creates artefacts not science.
Among other things, Lilly cooked the books. There was an excess of suicidal events on Prozac in their trials – statistically significant when you undo some regulation breaching maneuvers they undertook, with support from senior FDA officials. So the case reports and their trial data, clinical science, Evident Based Medicine and the new company EBM, map onto each other – as you might expect.
But if they didn’t map onto each other? The 1950s antidepressants treated melancholia, an illness that hugely increases the rate of suicides. SSRIs can’t treat melancholia; the SSRI trials were in mild depression, where the risk of suicide is minimal.
So in an RCT of these older drugs in melancholia we would expect the older drugs to save lives, even though doctors around 1960 were able to say these drugs can also cause suicide. In mild depression trials, the same older drugs, just like Prozac, show an excess suicidal acts.
Resolving discrepancies is how we do science. Lilly, however, were not in the business of being scientific. Their argument is a religious one – a dogmatic one – they forbid us to believe the evidence of our own senses.
The goal of marketing (propaganda) is not to increase sales of Prozac, it is to own the market. EBM was the moment Pharma took ownership of the market. It allowed them dictate what the risks, benefits and trade-offs of drugs are. It let them force us to live the lives they want us to live rather than engage in the business of helping us to live the lives we want to live.
This article transformed doctors into savages, who figure that even if it is evident Drug X has caused us Problems A-Z, silly me I must stop being taken in by anecdotes when the evidence says the opposite. If you insist your drug is causing a problem, if you continue to hold crazy ideas, I will have to detain you in a mental hospital.
Why did Richard Smith, the editor of the BMJ, allow this article to be published, when even his reviewers pointed to an excess of suicidal events on the drug and reading the small print would have made it clear the drug caused suicide? Smith was a mover in a groundswell that was leading to EBM. An article saying RCTs are more important than clinical science seemed like an opportunity to rope Pharma into something that would contain it.
And it was not totally unreasonable to think that a mindless algorithmic approach (RCTs) might make medicine more rational and deliver it from the vagaries of what some saw as arbitrary clinical judgement calls.
In 1999, Smith was presented with all the company documents pointing to the mistake he had made. He suggested that having created a problem perhaps BMJ should do something to solve it. He sabotaged efforts to do so.
He who would do a great evil must first persuade himself
he is doing a great good.
To be continued with A Dram of Beasley
This post links to